How did Mozart Compose?
The story commonly told about Mozart's genius—widely popularized by Peter Shaffer's play Amadeus, and later by his Oscar-winning film of the same name—portrays the composer as a divinely inspired idiot savant, who could see the piece whole in his imagination, like a painting or a statue, and who needed only time to write it down. While surviving historical documentation is by no means adequate to explaining Mozart's creativity, it does suggest a different narrative.
In the 1790s Constanze Mozart made a fateful decision about her late husband's musical manuscripts: those containing sketches or drafts of unrealized works would be kept for possible completion by others, while those containing sketches or drafts of completed works could be discarded. Nonetheless, enough sketch leaves of various sorts survive to enable us to see that, although Mozart never sketched to the extent that, for instance, Beethoven did, he used sketches in various ways throughout his career.
Cornell Library's Sketch in Mozart’s Hand
This original sketch leaf contains Mozart's resolution (recto, top) of one of the puzzle canons from Padre Martini's Storia della musica and a variety of jottings (verso, bottom). The verso of this leaf preserves a variety of Mozart’s jottings.
Giovanni Batista Martini, Storia della musica, 3 vols. (Bologna, 1757-81).
The 14-year-old Mozart studied counterpoint in Bologna with Martini, who presented him and his father with volumes 1 and 2 of his Storia (middle).
Sketch for the String Quartet in F major, K. 590
Facsimile of a sketch for the Minuet of the String Quartet in F major, K. 590 (1789). This manuscript belongs to the library of the Rimsky-Korsakov Conservatory, St. Petersburg.
Here, in a fine example of his sketching hand, Mozart has worked out an entire 42-measure minuet, establishing the continuity of ideas. The texture varies from one to four parts as Mozart indicates important accompanimental or contrapuntal details. The single measure on the eighth staff is probably a rejected idea for a trio.
Keyboard fragments
The writings about Mozart’s fragments conflate three distinct types: (1) a work that has been partly lost; (2) a work or movement begun and then broken off before the end was reached; and (3) a work drafted from beginning to end but with not all parts filled in. Mozart left more fragments of types 2 and 3 than any other major composer—approximately one fragment per four completed works.
[left] The splendid sonata movement in G minor, K. 590d represents type (2): Mozart put it aside after reaching the downbeat of measure 106. After his death another hand wrote four measures leading to the recapitulation, indicating that the first 21 measures could be taken over literally, and then providing 15 measures of retransition. The closing 33 measures are found only in the first edition, published in 1805.
[middle and right] The Rondo in A major, K. 386, is an example of types 1 and 3. Because the instrumentation was incomplete and the final leaf then missing (although since rediscovered), the autograph was dismembered by its owner, the English composer Sir William Sterndale Bennett (1816-1875), who gave away individual pages and parts of pages to autograph collectors.